Locally Optimum Detection of a Noise Model Based on Generalized Gaussian Distribution

In order to optimize signal detection in non-Gaussian environments, the work is addressed to provide realistic modeling of a generic noise probability density. The model depends on few parameters which can be estimated quickly and easily, and so general to be able to describe many kinds of noise such as symmetric or asymmetric. To this end, a new model is introduced, which derives from the generalized Gaussian function, and depends on there parameters: kurtosis, for representing variable sharpness, left variance and right variance, for describing deviation from symmetry. The model is applied in the design of a locally optimum detection test.